The majority of persons with chronic hepatitis B virus (HBV) infection live in resource-limited settings (RLS) such as China where 10-20% of the 780,000 HIV-infected patients are co-infected with HBV. Data from the Peking Union Medical College Hospital shows that half of all recent deaths in HIV-infected patients are attributable to liver failure, which is most commonly from HBV. Thus, HIV-HBV co-infection is a major public health problem in China and other RLS. Tenofovir as part of antiretroviral therapy (ART) is recommended for treatment of HIV-HBV co-infected patients in RLS, but this recommendation is not always followed due its cost and availability. Recent studies from RLS demonstrate that up to half of the HIV-HBV co-infected patients have low levels of HBV DNA; thus, therapy with lamivudine-based ART, which is much less expensive and widely available, may be efficacious in such co-infected patients. The novel hypothesis proposed in this study is that lamivudine-based ART has long-term efficacy in a subset of patients such as those with low HBV DNA. In Aim 1, we will compare the HBV virologic response and side effects in ~100 Chinese HIV-HBV co- infected patients who received lamivudine-based ART to ~100 who received tenofovir-based ART. We hypothesize that at low pre-treatment HBV DNA levels, these two treatment regimens will be similar but at higher levels, tenofovir will be superior. For this Aim, subjects will have serum tested for various HBV and HIV parameters every six months. We will also perform full genome HBV sequencing to look for mutations that exist prior to therapy and therefore may affect treatment outcomes or that emerge on therapy. We will also compare side effects of the regimens. Aim 2 compares the HBV-specific immunological response in lamivudine-based and tenofovir-based ART regimens. In this Aim, 20 subjects in each treatment group will have their HBV- specific T cell responses, including polyfunctionality, tested to a panel of HBV peptides prior to therapy and at 24, 48, and 96 weeks after therapy. This Aim will also use all 200 subjects to compare the cytokine production between these two treatment groups. We expect that the immunological response will not differ significantly between the groups especially in those who have low HBV DNA levels. Completion of the proposed Aims will provide much-needed data on the efficacy of lamivudine-based ART in a RLS. Such data are important since tenofovir is expensive and if a subset of patients, such as those with low HBV DNA, respond well to lamivudine, then treatment regimens in RLS can be individualized based upon the likelihood of response to lamivudine. This would allow better utilization of tenofovir, which is significantly more expensive, so that a greater number of patients can be treated for a given amount of money. This is a collaborative project between U.S. and China investigators where Dr. Thio, the U.S. PI, has expertise in HBV virology and Dr. Li, the China PI, has expertise in immunology.